From Raw Measurements to Final Result: A Reproducible qPCR Checklist From Raw Measurements to Final Result: A Reproducible qPCR Checklist A final result should be traceable backward. Starting from an effect estimate in a table, we should be able to identify the normalized values behind it, the reference measurements behind those values, the replicate decisions... Continue Reading →
Data Science – Medicine Application Project 3, Part 5
Do QC Exclusions Drive the Result? A qPCR Sensitivity Analysis # Do QC Exclusions Drive the Result? A qPCR Sensitivity Analysis Quality control protects an analysis from unreliable measurements. It can also change the sample size and, potentially, the conclusion. That creates a reasonable question: **Would we obtain the same result under another plausible replicate... Continue Reading →
Data Science – Medicine Application Project 3, Part 4
One Dataset, Three Normalizations: How the Estimated miRNA Effect Changes The target measurements have not changed. The study groups have not changed. The quality-control rule has not changed. Only the normalizer changes—and one apparent miRNA result changes from absent to present. This is why normalization cannot be treated as a cosmetic step at the end... Continue Reading →
Data Science – Medicine Application Project 3, Part 3
Reference RNA Stability: The Assumption Behind Every ΔCq Result A reference RNA is often treated as background machinery. Measure the target, subtract the reference, and move on. But the reference is inside every normalized value: **ΔCq = target Cq − reference Cq** If the reference changes between study groups, the normalized target changes with it.... Continue Reading →
Data Science – Medicine Application Project 3, Part 2
Technical Replicates in qPCR: Why Correlation Is Not Enough Two measurements can move together without being close enough to substitute for one another. That distinction is easy to miss in qPCR data. If replicate 1 and replicate 2 produce a correlation near 1, the assay may look almost perfect. But correlation answers a limited question:... Continue Reading →
Data Science – Medicine Application Project 3, Part 1
Before ΔCq: Building a Synthetic qPCR Dataset That Can Fail Honestly Most qPCR tutorials begin with a tidy spreadsheet: one row per sample, one expression value per target, and no explanation of how that number was produced. That is convenient, but it hides the part of the analysis where many important decisions are made. What... Continue Reading →
Data Science – Medicine Application Project 2 Part 5
Testing the result before trusting the story A statistical result is more useful when it survives reasonable analytical choices. This final part examines one uncertain variable and then turns the observed event rate into a prospective planning exercise. The main exposure is the operative bleb or bulla finding. Most records state that a lesion was... Continue Reading →
Data Science – Medicine Application Project 2 Part 4
Building a recurrence model that is evaluated out of sample The Cox model in Part 3 was designed to estimate adjusted associations. A prediction model has a different goal. It should produce useful probabilities for records that were not used to fit it. This analysis focuses on ipsilateral recurrence within 24 months. Records censored before... Continue Reading →
Data Science – Medicine Application Project 2 Part 3
A forest plot shows adjusted hazard ratios and 95% confidence intervals for ipsilateral recurrence.
Data Science – Medicine Application Project 2 Part 2
Comparing postoperative outcomes without losing clinical meaning The first part defined the cohort and checked its structure. The next step compares the two operative indication groups. Postoperative air leak, chest tube duration, and hospital stay are measured in days. These variables are bounded at zero and often have long right tails. A small number of... Continue Reading →